tuning the functional properties of lignocellulosic films

This is an electronic reprint of the original article.This reprint may differ from the original in pagination and typographic detail.

Powered by TCPDF (www.tcpdf.org)

This material is protected by copyright and other intellectual property rights, and duplication or sale of all or part of any of the repository collections is not permitted, except that material may be duplicated by you for your research use or educational purposes in electronic or print form. You must obtain permission for any other use. Electronic or print copies may not be offered, whether for sale or otherwise to anyone who is not an authorised user.

Gerbin, E.; Rivière, G. N.; Foulon, L.; Frapart, Y. M.; Cottyn, B.; Pernes, M.; Marcuello, C.;Godon, B.; Gainvors-Claisse, A.; Crônier, D.; Majira, A.; Österberg, M.; Kurek, B.;Baumberger, S.; Aguié-Béghin, V.Tuning the functional properties of lignocellulosic films by controlling the molecular andsupramolecular structure of lignin

Published in:International Journal of Biological Macromolecules

DOI:10.1016/j.ijbiomac.2021.03.081

Published: 30/06/2021

Document VersionPublisher's PDF, also known as Version of record

Published under the following license:CC BY-NC-ND

Please cite the original version:Gerbin, E., Rivière, G. N., Foulon, L., Frapart, Y. M., Cottyn, B., Pernes, M., Marcuello, C., Godon, B., Gainvors-Claisse, A., Crônier, D., Majira, A., Österberg, M., Kurek, B., Baumberger, S., & Aguié-Béghin, V. (2021). Tuningthe functional properties of lignocellulosic films by controlling the molecular and supramolecular structure oflignin. International Journal of Biological Macromolecules, 181, 136-149.https://doi.org/10.1016/j.ijbiomac.2021.03.081

https://doi.org/10.1016/j.ijbiomac.2021.03.081


International Journal of Biological Macromolecules 181 (2021) 136–149

Contents lists available at ScienceDirect

International Journal of Biological Macromolecules

j ourna l homepage: ht tp : / /www.e lsev ie r .com/ locate / i j b iomac

Tuning the functional properties of lignocellulosic films by controllingthe molecular and supramolecular structure of lignin

E. Gerbin a, G.N. Rivière b, L. Foulon a, Y.M. Frapart c, B. Cottyn d, M. Pernes a, C. Marcuello a, B. Godon a,A. Gainvors-Claisse a, D. Crônier a, A. Majira d, M. Österberg b, B. Kurek a, S. Baumberger d, V. Aguié-Béghin a,⁎a Université de Reims Champagne Ardenne, INRAE, FARE, UMR A 614, 51097 Reims, Franceb Aalto University, School of Chemical Engineering, Department of Bioproducts and Biosystems, P.O. Box 16300, FI-00076 Aalto, Espoo, Finlandc Laboratoire de Chimie et Biochimie Pharmacologiques et Toxicologiques—UMR CNRS 8601, Université de Paris, Franced Institut Jean-Pierre Bourgin, INRAE, AgroParisTech, Université Paris-Saclay, 78000 Versailles, France

⁎ Corresponding author at: Université de Reims ChamUMR A 614, 51097 Reims, France.

E-mail address: [email protected] (V. Aguié-Bé

https://doi.org/10.1016/j.ijbiomac.2021.03.0810141-8130/© 2021 The Authors. Published by Elsevier B.V

a b s t r a c t
a r t i c l e i n f o
Article history:Received 30 January 2021Received in revised form 13 March 2021Accepted 13 March 2021Available online 23 March 2021

Keywords:Protobind 1000Colloid lignin particles (CLP)Cellulose nanocompositeAntioxidant and antibacterial propertiesPhenoxy radicalsElectron paramagnetic resonance

This study investigated the relationships between ligninmolecular and supramolecular structures and their func-tional properties within cellulose-based solid matrix, used as a model biodegradable polymer carrier. Two typesof derivatives corresponding to distinct structuration levels were prepared from a single technical lignin sample(PB1000): phenol-enriched oligomer fractions and colloidal nanoparticles (CLP). The raw lignin and its deriva-tives were formulated with cellulose nanocrystals or nanofibrils to prepare films by chemical oxidation orpressure-assisted filtration. The films were tested for their water and lignin retention capacities, radical scaveng-ing capacity (RSC) and antimicrobial properties. A structural investigation was performed by infrared, electronparamagnetic resonance spectroscopy and microscopy. The composite morphology and performance were con-trolled by both the composition and structuration level of lignin. Phenol-enriched oligomers were the com-pounds most likely to interact with cellulose, leading to the smoothest film surface. Their RSC in film was 4- to6-fold higher than that of the other samples. The organization in CLP led to the lowest RSC but showed capacityto trap and stabilize phenoxy radicals. All filmswere effective against S. aureus (gram negative)whatever the lig-nin structure. The results show the possibility to tune the performances of these composites by exploiting ligninmulti-scale structure.© 2021 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license (http://

creativecommons.org/licenses/by-nc-nd/4.0/).

1. Introduction

A considerable effort has beenmadeworldwide to develop newma-terials from bioresources to replace petroleum-based plastics, whichcontribute to the acceleration of global warming and plastic pollution.To be competitive, the novelmaterialsmust combine several properties,such as low density, renewability, recyclability, processability and com-patibility with human health and the environment, and low productioncost [1]. Cellulose is the most abundant polymer found in forestry andagricultural biomass, and it meets all of the above requirements andcan thus be used as a building block for new biomaterials. Its potentialto produce high-performance nanocomposites filled with lignocellu-losic fibres (60–70%) has been demonstrated for forty years [2,3]. More-over, nanocellulose-based products have shown a wide range ofpotential applications, including engineered tissue for medical pur-poses, reinforcement fillers in packaging, sensors and flexible

pagne Ardenne, INRAE, FARE,

ghin).

. This is an open access article under

electronics, and protective coatings in optical devices. These productsinvolve either cellulose nanocrystals (CNC) prepared by controlledacid hydrolysis of amorphous domains [4] or cellulose nanofibrils(CNF) recovered from mechanical disintegration or steam explosion offibres [5,6]. Depending on the cellulose source (trees, plants, bacteria,algae, and tunicates) and the process used, the rod-like shape of thesenanomaterials varied in length from 100 nm to several μm, with an as-pect ratio ranging from 25 to higher than 1000 for CNC and CNF, respec-tively [7,8]. Both CNC and CNF represent attractive materials in thenanotechnology field [9,10]. Notably, their capacity to be aligned innanoassemblies is often exploited as an advantage [8,11].

Alongwith cellulose, lignin is a polymer of interest for biobasedma-terials. It is the second most abundant polymer (15–30%) in lignocellu-losic biomass and provides multifunctional compounds with highvalorization potential that are available as byproducts from lignocellu-lose processing in pulping or bioethanol production [12,13]. Lignin is aheterogeneous polymer crosslinked with cellulose and hemicellulosein plant cell walls. Several chemical and enzymatic processes using ad-vanced oxidative reactions have thus been developed to remove this re-calcitrant polymer from the cell wall and convert it into aromatic

the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).

http://crossmark.crossref.org/dialog/?doi=10.1016/j.ijbiomac.2021.03.081&domain=pdf

http://creativecommons.org/licenses/by-nc-nd/4.0/



mailto:[email protected]

Journal logo



http://www.sciencedirect.com/science/journal/

http://www.elsevier.com/locate/ijbiomac

E. Gerbin, G.N. Rivière, L. Foulon et al. International Journal of Biological Macromolecules 181 (2021) 136–149

products of industrial interest [14–16]. Consequently, the chemicalcharacteristics of the isolated raw material differ from those of nativelignin based on the type and severity of the removal process. Amongthose characteristics are the molar mass distribution, content of mono-mer units, pattern of intermonomer linkages (S, G, H), amount of func-tional groups (e.g., phenol, alcohol or methoxy groups) [14,17,18], andpresence of new groups such as sulfonate groups formed during theprocess [19]. In addition, the phenolic fraction is more or less contami-nated by carbohydrates, which are either solubilized along with ligninor present as a residual fraction after lignocellulose conversion[15,19–21]. Among the different industrial lignin production processes,the alkali (or soda) process has the advantage of not introducing anysulfur in the polymer backbone and preserving some of the lignin's na-tive structure, such as β-aryl ether bonds [12]. Moreover, as an aqueousprocess, it does not require organic solvents that are detrimental to theenvironment. One of themostwidely studied commercial soda lignins isPB1000 lignin (GreenValue Enterprises, USA) [17,22,23], which is pro-duced from a grass mixture and composed of several percent of freephenolic acids of potential interest. Combining solvent fractionationand ionic liquid treatment has allowed the depolymerization and de-methylation of this industrial lignin, thus yielding additional free phe-nols that promote antioxidant activity in solution [23]. In parallel,water-dispersible lignin nanoparticles (also named colloidal lignin par-ticles, CLP) were developed from kraft, alkaline and organosolv lignins[24]. These studies demonstrate that the lignin structure and supramo-lecular organization can be adjusted either by fractionation and depoly-merization processes or by promoting self-assembly. Moreover, all ofthese structures may optimize the valorization of the material inhigher-value applications, such as in nano- or microcarriers,supercapacitors for energy storage [25], high-performance carbon fibresfor microelectronics [26,27], highmechanical performance thermoplas-tics [28,29], multifunctional cellulose-based films (anti-UV, antireflec-tive, antioxidant in coated or casted films) [22,30–32], antimicrobial orantiviral nanoparticles [33,34] as flocculation agents or stabilizers inPickering emulsions [35–37], and water reducers for cement admix-tures. [38]

Lignin phenolics can be associated with a nanocellulose-based ma-trix to generate multifunctional composites that combine the mechani-cal resistance, transparency and barrier properties of nanocellulosewiththe hydrophobic, anti-UV, antioxidant and cross-linking properties ofphenolics [39]. More recently, lignin-carbohydrate complexes havebeen shown to be promising scavengers of reactive oxygen speciesand thus are highly sought-after in biomedical applications [40]. Inthese composite systems, the structural level of lignin might impactthe accessibility of its functional groups and consequently the proper-ties of the material, which has not been investigated thus far. Thus,the objective of this study was to explore tuning the lignin organizationlevel and determine the associated impacts on the functional propertiesof a polysaccharide-based matrix.

For this purpose, a nanocomposite system based on cellulosenanocrystals and nanofibrils was used as a model material, in whichPB1000 soda lignin was introduced in different structural forms andunder different conditions. The forms include raw lignin, an oligomerfraction, depolymerization products and colloidal lignin particles

Table 1Characteristics and solvent conditions of PB1000 and its derivatives before cellulose-based film

Lignin samples Mw g mol−1 Mn g mol−1 Pd PhO

PB1000a 1260 1015 6.4

F1CH40a 876 680 4.8F2CH40a 872 676 4.7CLPb – – –

a Information obtained from [23].b Colloidal particle of lignin prepared according to [34]; and Pd: polymerization degree estim

137

(CLP). The cellulose-lignin nanocomposites were prepared either afteroxidative reaction (using Fenton's reagent) prior to film formation orsimply by mechanical mixing with pressure-assisted filtration. Thesetwo procedures were selected to apply different crosslinking degreesof thewhole system. The paper focuses on two functional properties rel-evant to food packaging ormedical applications: radical scavenging andantibacterial properties. The structure of the composites was investi-gated through a combination of spectrometric, topographic andphysico-chemical sorption methods to assess the functional groupsand interactions between cellulose and lignin. This multiple approachallowed us to gain insights into the influence of both the initial molecu-lar structure and supramolecular structure of lignin duringfilm process-ing on the two functional properties of interest. The results show theimportance of analysing the lignin oligomer structure and adopted con-formation within the final material to enable the optimization of thefinal functional properties of the composite material.

2. Materials and methods

2.1. Materials

Technical grass lignin PB1000 was purchased from GreenValue En-terprises LLC (USA) [41]. Fractions F1CH40 and F2CH40 were obtainedby two combined solvent fractionation and ionic liquid treatment pro-cesses carried out on PB1000 according to a previously published ap-proach [23]. F1CH40 corresponds to the soluble PB1000 ethyl acetate(EA) extract after [HMIM]Br treatment (IL), and F2CH40 correspondsto the EA residue submitted to butanone extraction followed by IL treat-ment. Each IL treatment was performed in an Ace Pressure tube underan inert atmosphere with conventional heating at 110 °C. After 40min, the solid residue was filtered and washed with water (20 mL)and EA (20 mL). The filtrate was recovered, the layers were separated,and the aqueous layerwas extractedwith EA (2 times 20mL). The com-bined EA extracts were dried over MgSO4 and concentrated under re-duced pressure below 35 °C. The chemical characteristics of thePB1000 and EA fractions after [HMIM]Br treatment are summarized inTable 1. The lignin model compounds (LMCs) [39] included lignin frac-tions isolated from spruce wood (Picea abies) (FL-G) and maize stalks(Zea mays L.) (FL-GS), synthetic lignins (DHP-G) and a commercialdimer (guaiacylglycerol-β-guaiacyl ether, G2). The CLPs were preparedfrom PB1000 according to the procedure described by Rivière et al. [34].Briefly, 2 g of lignin was dissolved in acetone:water (3/1, v/v – 200 mL)for 3 h under magnetic stirring. The lignin solution was then filteredthrough a glass microfibre filter grade GF/F (Whatman, Maidstone,United Kingdom) before being poured into 500 mL of water under vig-orous stirring. The CLP dispersion was then dialyzed against deionizedwater to remove the residual acetone (the water was frequentlychanged). After 4 days, an aqueous CLP dispersion (c.a. 2 wt%) was ob-tained. Cellulose nanocrystals (CNC) were obtained from ramie fibres(Boehmeria nivea) after mild sulfuric acid hydrolysis as previously de-scribed [31]. Cellulose nanofibrils (CNF) were prepared from hardwoodkraft pulp fibres following the procedure described in [42]. After wash-ing in a sodium formula, the fibres were fibrillated using a type M-110P

preparation.

H mmol g−1 Solvent conditions Film matrix

free with cellulose

2.7 EtOH/Water6/4 (v/v)

EtOH/water3/7 (v/v)

CNC6.6 Dioxan/water

9/1 (v/v)Dioxan/water9/11 (v/v)8.3

– Water CNC or CNF

ated by usingmeanMwvalue of one phenylpropane unit (C9H9O2(OCH3)1–2), 196 gmol−1.


microfluidizer (Microfluidics, Newton, Massachusetts, U.S.A.) to obtaina CNF suspension (c.a. 2 wt%).

2.2. Particle size and ζ potential

The CLP dispersion was characterized using a Zetasizer Nano-ZS90(Malvern, United Kingdom). The particle size (referred to as the hydro-dynamic diameter) was determined via dynamic light scattering. The ζpotential was determined with a dip cell probe and calculated from theelectrophoretic mobility data using a Smoluchowski model. The resultsare reported as an average of three measurements. One millilitre of theCLP dispersionwas used for all measurementswith a concentration of c.a. 0.2 g L−1. Two different batches of CLP with similar size distributionsand ζ potentials were prepared using the same conditions, and their de-tailed characterization can be found in the supplementary document(Fig. S1). The ζ potential of CNC in suspension was characterizedunder the same conditions, and their particle size distributionwasmea-sured by atomic force microscopy (AFM) in a previous study [43].

2.3. Preparation of cellulose-lignin films

Cellulose-lignin filmswere prepared from CNC and CNF according totwo procedures: a chemical oxidation procedure with Fenton's reagent[22] or pressure-assisted filtration and drying at ambient temperature(25 °C) and 50% relative humidity (RH), respectively [32]. Mixtures ofCNC and lignin were prepared in an ethanol/water mixture (3/7, v/v)for PB1000, in dioxane/water (9/11, v/v) for F1CH40 and F2CH40 andin water for CLP to obtain nanocomposite films with 10 or 17wt% lignincontent. The addition of Fenton's reagent was adjusted to a FeSO4/H2O2

molar ratio of 10 (1mMFeSO4, 0.1mMH2O2 or 0.3mMFeSO4, 0.03mMH2O2) at pH 3with diluted sulfuric acid to ensure amild oxidation reac-tion in each suspension. Themixturewas stirred for 2 h at ambient tem-perature before film preparation by casting onto a horizontal PTFE plateand drying overnight under ambient air and relative humidity. The vol-ume of suspension poured in the PTFE plate was adjusted for each mix-ture to obtain an average film thickness (CNC-lignin) on the order of 27± 3 μm, which was measured with a dual-thickness gauge (KäferGmbH, Villigen, Germany). The thickness of the nanocomposite filmsprepared from CNF averaged 80 ± 3 μm. All cellulose-based filmswere stored under controlled conditions (23 °C, 50% RH) in the dark be-fore testing. In parallel, supported thin filmswere produced by applying300 μL of each cellulose-lignin mixture to clean quartz slides (3.1 cm2)and then drying the slides under ambient conditions overnight. Allfilms were conditioned according to their characterizations.

2.4. Structural and spectroscopic characterizations

2.4.1. Fourier transform infra-red (FTIR) spectrophotometryFTIR spectra were recorded for each initial lignin and the films using

a Nicolet 6700 spectrophotometer (Thermo Scientific, U.S.A.). The tab-lets were prepared by mixing 200 mg of spectroscopic grade KBr with3mg of lignin or crushedfilm. In the case of CNF-basedfilms, the spectrawere recorded by attenuated total reflectance. Each spectrum repre-sented a record of 32 scans at a resolution of 4 cm−1 from 400 to 4000cm−1, and we performed background subtraction, baseline correctionand then normalization with respect to the area under the spectrum.

2.4.2. Quantitative 31P NMRDerivatization of the LMC samples with 2-chloro-4,4,5,5-

tetramethyl-1,3,2-dioxaphospholane (Sigma-Aldrich, France) was per-formed according to [44]. NMR spectra were acquired on a BrukerBiospin Avance III 500MHz spectrometer. A total of 256 scanswere per-formed with a delay time of 6 s between successive pulses. The spectrawere processed using Topspin 3.1 from Bruker Biospin. All spectra weremanually phase-corrected and calibrated with the observed signal fromthe reaction product between water and 2-chloro-4,4,5,5-tetramethyl-

138

1,3,2-dioxaphospholane in pyridine/CDCl3 at 132.2 ppm. Signals wereassigned by comparison with the 31P NMR chemical shift range as indi-cated in [22].

2.4.3. Electron paramagnetic resonance (EPR)Spectra were recorded at 20 °C using a Bruker 540 EPR spectrometer

operating in the X-band (9.85 GHz) under the following conditions:modulation frequency, 100 kHz;modulation amplitude, 10G; time con-stant, 327.6 ms; conversion time, 327.91ms; and microwave power, 10mW. Data acquisition, processing, and double integration were per-formed using Bruker software. Approximately 20 mg of crushed filmwas added to EPR tubes (Wilmad, 707-SQ-100 M). The amplitude ofthe signal was calculated based on the difference between the maxi-mum and the minimum of the spectral signal, which was possible be-cause the EPR signal was symmetric.

2.4.4. Atomic force microscopy (AFM)AFM images of cellulose-based filmswere acquired on aMultimode-

8 AFM setup (Bruker, Santa Barbara, CA) in PeakForce tapping mode inair. Images were taken with V-shaped silicon nitride ScanAsyst-air AFMprobes (nominal spring constant of 0.4 N m−1). Prior to AFM image ac-quisition, the AFM tip was calibrated. The deflection sensitivity of theAFM probe was achieved by recording at least three force-distancecurves on a stiff sapphire flat surface. The spring constant was deter-mined by the thermal tuningmethod [45]. The nominal sharp tip radiusis 2 nm, which minimizes nondesirable broadening effects [46]. AFMimages were acquired with a vertical tip oscillation frequency and ac-quisition rate of 1 kHz and 0.4–0.6 Hz, respectively. The image resolu-tion was defined as 512 × 512 pixels/line. Several images were takenin different scan areas ranging from 400 μm2 to 25 μm2 to determinethe roughness parameters of the films (Ra) and provide an overviewof lignin nanoparticle distribution on each sample. Ra is defined as thearithmetic average of the absolute values of the surface height devia-tions measured from the mean plane. Roughness parameters will aidin a more complete overview of how lignin nanoparticles are dispersedon the cellulose film surface. Moreover, to precisely quantify the dimen-sions of lignin features, cross-section profileswere produced (N=300).Histograms were fitted by Gaussian distribution using OriginPro 8.0software. Roughness film parameters of the entire AFM images andcross-section lignin profiles were rendered by Nanoscope Analysis 1.8software.

2.4.5. Dynamic vapor sorption (DVS)A gravimetric sorption analyser (IGA, Intelligent Gravimetric

Analyser, Hiden Isochema Ltd.) was used tomeasure thewater sorptionisotherm of each film [47]. An ultrasensitive microbalance located in athermostatically controlled chamber was used to measure changes infilm mass (approximately 4 mg) as low as 0.1 mg at 20 °C and at RHvalues from10 to 90%. Thefilmwashydrated stepwise via 10% RH incre-ments during the water vapor sorption and desorption phases, and themass was recorded until equilibrium at each stage. The water content(Cw) was calculated according to Eq. (1):

Cw ¼ meq−md

md� 100 ð1Þ

where meq is the mass of the sample in the equilibrium state and md isthemass of the dry samplemeasured after total drying under a dry flowof nitrogen. Sorption isothermswere analyzed by using the PARKmodelaccording to a previous analysis [39,47].

2.5. Antioxidant activity determination

2.5.1. Soluble radical scavenging activity of lignin in solutionThe soluble radical scavenging activity of lignin samples was evalu-

ated both in absolute ethanol solution and in water by measuring


their reactivity towards the stable soluble radicals (R•) 2,2-diphényl-1-picrylhydrazyle (DPPH•) [48] and 2,2′-azino-bis(3-ethylbenzothiazo-line-6-sulfonic acid radical cation (ABTS•+) [49]. Each lignin samplewas solubilized in ethanol/water (6/4, v/v) or dioxane/water (9/1, v/v)to obtain concentrations between 0.25 and 0.5mgmL−1. The dispersionwas stirred until total solubilization before its radical scavenging activ-ity was measured.

2.5.1.1. DPPH• test. In a quartz cuvette, 77 μL of lignin solutionwas addedto 3 mL of 60 μMDPPH• in absolute ethanol solution. The absorbance at515 nmwas monitored using a UV–Vis double-beam spectrophotome-ter (Shimatsu, Japan) until reaching a plateau. A blank was preparedunder the same conditions using 77 μL of solvent instead of the ligninsample. The kinetics of radical scavenging were analyzed from at least5 solutions in the concentration range mentioned above by calculatingthe difference between the absorbance of the sample and blank solutionevery 5 and 10 min. The percentage of residual DPPH• was calculatedand plotted vs the concentration of soluble lignin in solution.

2.5.1.2. ABTS•+ test. ABTS was dissolved in water to a concentration of 7mM. ABTS•+ was produced by reacting ABTS stock solution with a finalconcentration (2.45 mM) of potassium persulfate in the dark at roomtemperature overnight before use [50]. The solution was diluted untilachieving absorbance at 734 nm of 0.7 (± 0.02), which correspondedto a final ABTS•+ concentration of 35 μM measured at 415 nm, with amolar extinction coefficient ε=3.6 × 104mol−1 L cm−1. In a quartz cu-vette, 30 μL of CLP solution was added to 3 mL of ABTS•+ solution andthe absorbance was monitored at 734 nm under the same conditionsas described for the DPPH• test.

For both methods, the efficient concentration of antioxidant mole-cules needed to reduce 50% of the initial DPPH• or ABTS•+ (EC50sol)was determined from the respective curve and expressed in g lignin/mol R•.

2.5.2. Scavenging activity of lignin in cellulose-based films towards solubleradicals, R•

The soluble radical scavenging activity of CNC-lignin and CNF-ligninfilms was evaluated by measuring their reactivity towards both DPPH•andABTS•+ stable free radicals according to a previous study [22]. Pieceswith film masses ranging from 0.1 to 4 mg were cut and immersed in 3mL absolute ethanol DPPH• (60 μM)orwater ABTS•+ (35 μM) in a quartzcuvette. The absorbance (515 nm for DPPH• and 734 nm for ABTS•+)was measured until reaching a plateau vs a quartz cuvette withoutfilm to determine the EC50film expressed in g lignin/mol R• as tests of lig-nin in solution. A similarmeasurementwas carried out on pure cellulosefilms as a control. A comparison of EC50sol and EC50film was performedafter the expression of both parameters in g lignin/mol R•. To distinguishbetween the free radical scavenging activity of lignin films to the ligninethanol- or lignin water-extractable fractions, similar measurementswere made towards DPPH• or ABTS•+ after 360 min of infusion of filmin absolute ethanol (3 mL) or water (3 mL), respectively, as in [22].The supernatant was mixed with 30 μL of 60 μM DPPH• or 35 μMABTS•+, and the absorbance was measured until the plateau wasreached against the respective blank. The scavenging activity of the re-sidual film towards DPPH• or ABTS•+ was evaluated as previously de-scribed. The percentage of antioxidant activity of lignin released inethanol or water media was calculated at the steady-state achievementof each kinetic. All the experiments were carried out in duplicate.

2.6. Antibacterial activity

The antibacterial activity of cellulose-lignin films was evaluatedagainst Escherichia coli (CIP 54–127) and Staphylococcus aureus (CIP53–154). The overnight culture prepared in nutrient broth (37 °C, 150rpm, 16 h) was diluted (108 CFU/mL) before aseptic transfer of 250 μLto petri dishes containing composite films (free film sections (5 mm ×

139

5 mm) or supported coating film on quartz slides (3.1 cm2)). The filmswere incubated at 37 °C for 1 to 3 h under static conditions. Similarly,empty petri dishes and CNC and CNF film sections without lignin wereincubated with the same bacterial suspension at the same temperatureand time and used as positive controls. In the particular case of CNF-based films, circular sections (3.1 cm2) were used. At the end of the in-cubation period, the inhibitory effect was estimated by measuring thetotal viable cell count after plating 1 mL of the inoculum samples onplate count agar with serial dilutions from 10−1 to 10−8 (37 °C, 24 h).All tests were performed in triplicate on one-week-old films. Growthfactor reduction is equivalent to the log reduction (Log R), and it was es-timated using Eq. (2):

Log R ¼ Log A½ �−Log B½ � ð2Þ

where “A” is the mean value of three bacterial concentrations withoutfilm and “B” is the mean value of three bacterial concentrations in con-tact with cellulose-lignin film after the same time of incubation. A Log Rvalue of 1 is equivalent to a 10-fold reduction or a 90% reduction of thepopulation. A Log R value of 4 is equivalent to a 104-fold reduction or a99.99% reduction of the population.

3. Results and discussion

The antioxidant and antibacterial properties of alkaline lignin incellulose-based films were investigated as a function of the molecularand supramolecular lignin structures and the process used for its incor-poration into the cellulosic network. For that, two gradual scales werechosen (Fig. 1). The first one was based on the molecular weight and/or the structural-morphology state of lignin before its use: monomer-dimer of lignin (CA), raw lignin sample solubilized in ethanol/water(6/4, v/v) solvent, two fractions of PB1000 treated by ionic liquid andsolubilized in dioxane/water (9/1, v/v), or spherical CLP dispersed inwater (Table 1). The second scale was established by the organizationdegree of the film brought by using two different nanocelluloses (CNCand CNF), which implies a variation in film thickness.

3.1. Characteristics of PB1000 and its different derivatives

PB1000 is composed of 50% guaiacyl (G) units, 49% syringyl (S) unitsand 1% p-hydroxyphenyl (H) units (in moles) with low contents of p-coumaric and ferulic acids (~0.5 wt%) [23]. These G, S and H units aremainly linked through resistant carbon‑carbon bonds (5–5, β-5, β-β)and contain a small proportion of units only linked by aryl-aryl etherbonds (β-O-4), as indicated by its thioacidolysis yield of 83 μmol g−1,which is ten times lower than that of native grass lignins. According toits apparent average molar mass (Mw), the average polymerization de-gree of PB1000 was estimated to be 6.4, assuming 196 g mol−1 as themean unit molar mass value of lignin calculated from its empirical for-mula (Table 1) [51]. The free phenolic group content (2.7 mmol g−1)was similar to that of other technical lignins [17] and consistent withthe low content of (β-O-4) linkages. The water-soluble fraction (lessthan 2%, w/w) was composed of p-OH benzoic and p-coumaric acid, p-OH syringaldehyde and dimers/trimers of S and G units (data notshown).

The two fractions F1CH40 and F2CH40 produced by the IL treatmentof PB1000 were previously characterized [23]. They showed the samemolar mass distributions (Mn 676–680 g mol−1 and Mw 872–876 gmol−1) and corresponded on average to tetra- or pentamers. F1CH40and F2CH40 were both typified by enrichment in free phenols (6.6and 8.3mol g−1, respectively) (Table 1). Moreover, they both containedphenolic monomers, mainly acidolysis ketones and acetosyringone thatwas demethylated once or twice [23]. These two fractions providedtools to assess the influence of lignin phenol content on the system.

CLP prepared from PB1000 had a very low polydispersity index(<0.15) and an average hydrodynamic diameter of 114.7 ± 5.2 nm.

Fig. 1. Strategies to prepare cellulose-lignin-based films: mild chemical oxidation with Fenton reagent or filtration and hot drying pressure procedures using various phenolic structuresranging from oligomers to nanoparticles consisting of multiple oligomers. Images of coating and casting CNC films with 17 wt% PB1000 or CLP and CNF film with 10 wt% CLP.


Theywere easily dispersed inwater to form stable aqueous suspensionsas such or in the presence of cellulose nanocrystals because of their neg-ative ζ potentials of−28.2 ± 4.3 mV and −50.1 ± 0.9 mV for CLP andCNC, respectively, in water. The CLP were used to assess the influenceof the PB1000 supramolecular organization on the functionality of thecomposites. Indeed, since the CLP recovery yield exceeded 85%, itcould be assumed that the chemical composition of CLPwas representa-tive of the initial PB1000 sample except for the possible loss of thewater-soluble fraction. Table 1 summarizes the lignin sample character-istics, solvent conditions and cellulose nanoparticle type used for pre-paring the films and performing radical scavenging tests in solutionsor in cellulose films.

3.2. Characteristics of uniform and stable lignin-cellulose suspensions be-fore film preparation

A previous study demonstrated the possibility of graftingdehydropolymers (DHPs) and organosolv lignin oligomers onto CNCby mild chemical oxidation with Fenton's reagent in aqueous dioxanesolvent [22,30] or by in situ polymerization of aromatic precursors of lig-nin (coniferyl alcohol) in water [39,47]. These treatments conferrednew properties to the cellulosic materials, such as UV protection andtransparency, antioxidant properties, and stable organic radical poly-mers (ORPs) [22,39]. A similar procedure was applied here tosuspensions containing PB1000 or its derivatives using appropriate sol-vents (Table 1). The addition of Fenton's reagent to CNC-ligninmixturescontaining 10 or 17 wt% lignin led to the formation of colloidal suspen-sions that were stable at room temperature for several months (Fig. 1,photographs). Indeed, lignin oligomers were previously shown toform nodular structures when the monomers were drop-by-drop

140

polymerized in water or in aqueous solution of xylan [52] or uponself-assembly when water was added into THF/lignin solution [53].The uniform yellow colour of all the CNC-lignin suspensions and thelack of precipitation demonstrated the colloidal stability of each systemafter chemical oxidative reaction and suggested a homogeneous distri-bution of aromatic molecules or nanoparticles onto the cellulose nano-rods or fibrils. Both chemical oxidation before the evaporation stepand physical mixing before the pressure-assisted filtration processesled to composite films with good washing resistance. This finding sug-gested good cohesion between lignin and CNC or CNF surfaces, eitherthrough noncovalent interactions, such as hydrogen bonds and hydro-phobic interactions, or through covalent grafting via oxidation, whichis consistent with previous studies [22,32].

3.3. Nanoscale surface morphology of cellulose-lignin films

The surface morphology at the nanoscale of the different films wasanalyzed by AFM. The AFM images provided three-dimensional infor-mation, including the surface roughness values of the films. AFM imagesrepresentative of each CNC-based film at 45% RH are depicted in Fig. 2.CNC-F1CH40 andCNC-F2CH40films (Fig. 2A andB) showed topographyimages (5 μm × 5 μm) with roughness (Ra) varying from 10 to 37 nm.CNC rods with a mean length value of 180.9 ± 12.5 nm [54] are notwell distinguished due to the embedding phenomena by phenolic mol-ecules more or less modified during the oxidative reaction. Only a fewnodules could also be observed, linked perhaps to their higher phenolcontent (Table 1), which favours the interaction of these lignin deriva-tives with cellulose by hydrogen bonds. The AFM image of the CNC-PB1000 film showed on the contrary high amounts of nodules on thecellulose matrix with mean diameters of 383.6 nm ± 76.2 nm (Fig. 2C

Fig. 2. AFM topography analysis and size distribution of lignin nanoparticles in cellulose-based films. (A) CNC-F1CH40, (B) CNC-F2CH40, (C) CNC-PB1000, (D) CNC-CLP, and (E) CNF-CLP.Scan sizes for all AFM images are 20 μm×20 μm. The inset figures represent zoomed images of 5 μm×5 μm. Size distribution of lignin nanoparticles in CNC-PB1000 (D'), CNC-CLP(E') andCNF-CLP(F') (N = 300).


and C'). The Ra roughness of approximately 43 nm observed for the 5μm× 5 μm image was higher than that observed for the other samples.These globular structures have beenpreviously observed after oxidationof coniferyl alcohol and lignin oligomers with Fenton's reagent [31,39].They are most often randomly distributed throughout the film, andtheir formation seems to be initiated in the aqueous suspension in con-tact with hydrophilic and hydrophobic faces of cellulose nanocrystalrods under moderate stirring, and they tend to grow in size during thedrying step. Some nodules display “donuts” or an annular morphology(Fig. 2C') that is similar to the morphology observed with spray dryingof CNC suspensions by hydrophobic lignin [55] or in organometallicnanocomposites based on zinc or iron oxides [56,57]. According tothese previous works, the appearance of “donuts” could be explainedby the hydrophobicity of lignin, which increases with the water evapo-ration kinetics during the drying step of the film. This process couldcause a higher concentration of lignin in the nanoparticle centre withlocal high hydrophobicity due to the local faster evaporation of waterthan that on the surface of the crystalline faces of CNC. The presenceof metal ions, such as Fe2+ and Fe3+ from Fenton's reagent, can also co-ordinate strong interactions between lignin and the surface of CNC, suchas van der Waals bonds and hydrogen bonds, and induce particular as-sembly in the form of donut-like structures. The size distribution ofthese annular structures could also depend on the rate ofwater additionto the lignin solution, and itwas inversely associatedwith the rate of ad-dition of lignin solution to an excess of water [53]. Finally, according tothe AFM image analysis, the CNC-CLP films showed a uniform distribu-tion of nodules without donut-like structures, and they had a diameter

141

range between 100 nm and 150 nm and a mean value of 127.2 nm ±12.1 nm, which was close to that measured for the CLP suspension inwater by DLS 110.8 nm ± 0.7 nm (Fig. 2D and D', SI1). The roughnessof the CNC-CLP film was 14 nm ± 1 nm and lower than that of theCNC-PB1000 films. Thus, the mild oxidative reaction with Fenton's re-agent in the CNC suspension did not significantly modify the CLP mor-phology. In addition, the CLP supramolecular structure appeared to bepreserved in the CNF matrix, which is consistent with a previous study[58]. The structure was even preserved after a pressure-assisted filtra-tion process, and the mean diameter was 194.9 nm ± 32.7 nm andthe roughness was 24 nm (Fig. 2E and E'). For both studies, the ligninnanoparticles seem to be uniformly adsorbed on the CNF surfaces. Thehigher roughness values of the CNF-based films measured in our studycan be explained by the use of native CNF without chemical surfacemodifications and lignin nanoparticles with a 2-fold to 5-fold highermean size (CLP, 114 nm ± 5.2 nm) instead of TEMPO-oxidized CNFmixed with lignin nanoparticles prepared from mixed andprehydrolysed hardwood (nanoparticle sizes 10–50 nm). Thus, theCLP does not entirely penetrate the nanofibril network during theprocess.

3.4. Assessment of water accessibility of the films

The water sorption isotherms of cellulose-lignin films obtained bydynamic vapor sorption (DVS) measurements were used to performan indirect assessment of the intramolecular voids left between ligninnodules and cellulose nanorods or nanofibrils. These voids can form


nano- or microcavities that are accessible by water molecules (Fig. 3).Both the oxidative treatment and the pressure-assisted filtration pro-cess induced some linkages between lignin and the CNC surface, thus in-creasing thewater retention capacity of thefilms compared to the initialnontreated CNC and CNF films [22,32]. All samples exhibited sigmoidalcurves during water sorption and desorption in a 10–90% relative hu-midity range. This phenomenon, called hysteresis, was previously ob-served and followed a similar trend [22,32], and it is associated withthe deformation of the films induced after swelling during the watersorption steps. For the films composed of raw PB1000 and CLP, thewater content increased until approximately 25% when the RH in-creased from 10 to 90% by the formation of water sorption sites on thesurface of the polymer chain intertwined inside the film (representedby the AL factor between 0 and 20% RH), the linear sorption of water(represented by KH between 20 and 60%) and finally the formation ofwater molecule aggregates (represented by Ka and n) according to thePark model (Fig. 3, Table insert). No significant differences were ob-served between the Park parameter values for both PB1000 and CLP in-corporated at 10 wt% in CNC-based films, except for CLP mixed withcellulose nanofibrils, where AL was the highest value (0.10 gwater · g−1

dry

matter) of the film range and Ka was the lowest (0.25 gwater · g−1dry matter),

with a n value of 9.5. This observation suggested that the nucleationsites of water aggregates in the CNF-based film prepared by thepressure-assisted filtration process were smaller than those in filmsprepared by oxidation with Fenton's reagent. In the same manner, thehysteresis profile (D\\S,%) showed that water molecules at RH < 20%and water aggregates at RH > 60% could be trapped in the network,even after several hours of desorption, until an equilibrium state wasreached, i.e., 0.5% and 2.5–3 % water content, respectively (Fig. 3A in-sert). In contrast, for F1CH40- and F2CH40-derived lignin, the watercontent in the RH range at the end of the isotherm (RH > 60%) washigher than the water content of PB1000-containing films and then in-creased to reach the highest values for the Ka and n parameters (0.43gwater · g−1

dry matter for F1CH40 and F2CH40, with n values of approxi-mately 9.8 and 10.1, respectively (Fig. 3B)). This result suggested thatthenumber and/or size of thewater aggregates increased for these sam-ples andwere less easily desorbed from the network during the desorp-tion step in comparisonwith PB1000 (Fig. 3B insert). These phenomenacould be explained by the increased content of phenolic groups, whichare more likely to interact with the hydroxyl groups of cellulose via hy-drogen bonds, and the sorption of water molecules, which can become

Fig. 3.Water sorption isotherms and hysteresis (D\\S, % insert) of CNC- and CNF-based films wcomparison for PB1000 and derivatives after IL treatment (F1CH40 and F2CH40) in CNC films. Dthe Park equation described in [47]. The calculated deviation modulus E is inferior to 2%.

142

the nucleation sites of water aggregates trapped inside the network athigh RH. In conclusion, all Park parameter values indicated that the af-finity of cellulose-lignin films towardswatermolecules and their capac-ity to trap water aggregates were dependent not only on thepolymerization degree of the lignin oligomer but also on the phenolicgroup content. This dependency could be explained by the influenceof both parameters on the formation of nodular structures of varioussizes embedded in the cellulosematrix. To better understand the chem-ical structure of all films, FTIR measurements were performed on thesame samples to identify variations in functional groups before andafter oxidative or physical film processing.

3.5. Influence of film processing on lignin functional groups

FTIR analysis of the isolated lignin samples before film preparation(Fig. 4A) indicated that PB1000 and CLP had similar spectra, thusconfirming that CLP retained the main structural characteristics ofPB1000. The bands at 1708 cm−1 and 1215 cm−1 were assigned respec-tively to conjugated aldehydes and carboxylic acids (C=O) and to aryl-ether groups (C\\O); the bands at 1600, 1515, 1425 and 1269 cm−1

were assigned to aromatic skeleton vibrations; the band at 1462 cm−1

was assigned to C\\H stretching in the –CH3 and –CH2- groups; andthe bands at 1115 and 1032 cm−1 were assigned to C_O stretching vi-brations and aromatic CH- in plane deformation, respectively [59,60].The only differences between PB1000 and CLP occurred in the intensitydecrease of the bands at 1328 cm−1 (C\\O deformation in syringylunits), 1515 cm−1 and 1608 cm−1 (aromatic skeleton) and the disap-pearance of the band at 1315 cm−1 (CH2 rocking/wagging). These dif-ferences were consistent with the location of such apolar structures inthe core of the nanoparticles, which is consistent with previous NMRanalyses [61]. The spectra of the two fractions F1CH40 and F2CH40showed major differences with PB1000 and CLP [26,60,62]: increasedintensity of the OH stretching bands between 3100 and 3300 cm−1

and decreased intensity at 1708 cm−1 (C_O ester groups), at 1515and 1425 cm−1relative to the aromatic ring bonds), at 2915 cm−1 and1460 cm−1 relative to the C\\H stretching in the methoxy groups andat 1325 cm−1, which corresponded to the resonance of the C\\Ostretching of the S units [26]. The absorption maxima of the etherbond band at 1211 cm−1 for PB1000 and CLP shifted to lowerwavenumbers at 1200 cm−1 associated with phenolic OH deformationfor the two derivatives (Fig. 4A). These differences are fully consistent

ith 10 wt% lignin. (A) Data comparison for PB1000 and CLP in CNC and CNF films. (B) Data, desorption data; S, sorption data. Tables: respective Park parameters calculated by fitting

Fig. 4A. FTIR spectra of raw and derived PB1000 lignin samples. The vertical lines indicate the specific bands of lignin oligomers.


with the fact that EA fractions of PB1000 after [HMIM]Br treatmentarose from the cleavage of ether bonds and yielded additional free phe-nols in all the fractions (F1CH40 and F2CH40, Table 1) because of depo-lymerization and demethylation [23].

After their respective introduction in CNC-based films, the PB1000and CLP spectra maintained their main characteristic bands without

Fig. 4B. FTIR spectra of raw andderived PB1000 lignin samples in CNC-basedfilmswith 17wt% lwith CNC-CLP film with 10 wt% lignin content and measured by Attenuated Total Reflectance.

143

any variation (Fig. 4B), except for the shifting of the band at 1708cm−1 to 1716 cm−1 (C_O stretching in unconjugated ketones, car-bonyls and esters), whichwas caused by oxidationwith the Fenton's re-agent, and the band at 1515 cm−1, whichwas related to the vibration ofaromatic rings. The intensity of the band at 1716 cm−1 increased forF1CH40 and F2CH40, and the intensity of the band at 1515 cm−1

ignin content andmeasured onKBr tablets. Insert: IR spectra of CNF-CLPfilm in comparison

Fig. 5.Antioxidant properties of raw and derived PB1000 in solution (A) and in CNC-basedfilms (B) towards DPPH• and ABTS•+.


decreased in comparison with the same bands observed in the PB1000and CLP spectra. The ratio between the intensity (I) of these twobands I1716/I1515 was higher for F1CH40 and F2CH40 than for PB1000and CLP in the CNC-based films (1.1 and 0.7, respectively). This findingcould explain the slight generation of phenolic esters or ketones afteroxidative reaction using Fenton's reagent for these two PB1000 deriva-tives with a higher content of phenolic groups after the combined pro-cess (Table 1). Nevertheless, the F1CH40 and F2CH40 spectra weresimilar to those of the PB1000 and CLP-based films between 1060 and950 cm−1, suggesting that IR absorption was mainly governed by thecellulose network characterized by these bands (Fig. 4B). This last resultwas evenmore apparentwith the CNF-basedfilmprepared by pressure-assisted filtration and drying at ambient temperature (Fig. 4B insert).Furthermore, no additional band was observed at 1740 cm−1, which isinconsistent with previous studies that used higher concentrations ofhydrogen peroxide [30,39] or peroxidase [47].

3.6. Influence of ligninmolecular and supramolecular structures on the rad-ical scavenging capacity (RSC) of the material

To assess the influence of molecular and supramolecular structuresof lignin on their radical scavenging properties and thus their antioxi-dant potential, RSC measurements were performed on three types ofsystems with increasing organization level: isolated lignin in solution,CLP aqueous suspension, and lignin in cellulosic solid films.

3.6.1. RSC in solution/suspension without celluloseThe radical scavenging capacity (RSC) of all lignin samples was

tested towards DPPH• and ABTS•+ in absolute ethanol and water with-out cellulose by measuring the EC50sol parameter, which is defined asthe antioxidant concentration required to reduce 50% of the initial freeradicals [22]. This parameterwas determined from Fig. 5 and representsthe percentage of radical species reduced versus the lignin content insolution (Fig. 5A) or in film (Fig. 5B) per molar content of free radicals(R•) in medium (absolute ethanol for DPPH• and water for ABTS•+). Alarger EC50sol value corresponds to a lower RSC. PB1000 solubilized inaqueous organic solvent (ethanol/water (6/4, v/v)) showed a similarRSC towards DPPH• as CLP dispersed in water, and it had an averageEC50sol value of 0.87 ± 0.08 g of lignin per mole of DPPH• (correspond-ing to lignin bulk concentration of 0.33 ± 0.03 g L−1). The nanoparticlestructure of CLP is not stable in absolute ethanol solution and loses itsnodular structure; thus, similar EC50sol values were obtained for CLPand raw PB1000, which could be explained by the recovery of the initialdissolved state after dilution of CLP in the ethanolic solution imple-mented for the test (Fig. 5A). This finding also demonstrates that thePB1000 oligomerswere not chemicallymodified during the preparationof CLP. To avoid this effect of dissolution, tests were also performed forABTS•+ inwater, wherein the CLP structurewas preserved. Both PB1000and CLP showed lower EC50sol values than the DPPH• tests, and thisfinding reflected the more efficient scavenging process in aqueousABTS•+ than in ethanolic DPPH• solutions. This increased efficiencywas less pronounced for CLP than for PB1000 (0.48 ± 0.04 g mol–1 forCLP and 0.36 ± 0.04 g mol–1 for PB1000), suggesting that the organiza-tion into nanoparticles limited the access of the radicals to some phenolgroups. This limited access could be explained by the presence of phenolgroups in the core of the particles (25%, according to the EC50sol differ-ence between PB1000 and CLP systems) and/or to the possible stericfactors that control the reactions [63]. Indeed, the reaction becomes im-peded as the structural complexity and size of the molecule increasesdue to the additional rings that interfere with access to ABTS•+ andgreater number of DPPH• sites that are landlocked between two ben-zene rings (Fig. 5A insert). Nevertheless, most of the RSC of PB1000was preserved, even after self-assembly into CLP. F1CH40 and F2CH40were previously shown to have antioxidant properties that competewith coniferyl alcohol, ferulic acid and BHT in solution towards DPPH•

[23]. These fractions were not suitable for the ABTS•+ test due to their

144

tendency to aggregate in total aqueous solution. Thus, consistent withtheir 2- to 3-fold higher phenol group contents, these fractions exhib-ited a 3-fold lower EC50sol, which was determined through the DPPH•

tests (0.30± 0.01 gmol−1 corresponding to a lignin bulk concentrationof 0.11 ± 0.005 g L−1).

3.6.2. RSC in cellulosic filmThe RSC towards DPPH• and ABTS•+ were measured for all CNC-

lignin films, and the corresponding EC50film (efficient lignin concentra-tion in film required to reduce 50% of the radicals) was calculated(Fig. 5B). This parameter was measured by selecting the film massrange (g) to obtain a linear relation between the percentage of reducedradicals and the ratio g lignin/mol. R•. Since no antioxidant properties ofpure CNC or CNF film could be detected, these EC50film values werecompared with the EC50sol values for each lignin sample (Fig. 5A) to as-sess the influence of the process of incorporating lignin into the film.Moreover, the RSC of the films was related to the presence of phenolgroups, either in the film or at the surface of the material or carried byphenolic compounds released in the radical solution during the test[22]. As for EC50sol, the EC50film values varied according to the structureof the lignin samples andwere lower with the aqueous ABTS•+ solutionthan with ethanolic DPPH• solution. Thus, for the CNC-PB1000 film, theEC50film valuewas 3-fold lower for ABTS•+ thanDPPH• (1.21±0.04 ver-sus 3.98 ± 0.05 g mol−1) and 2.5-fold lower for the CNC-CLP film (2.22± 0.04 versus 6.19 ± 0.05 g mol−1). Expressed in mg of tannic acidequivalent (TAE), the RSC of the CNF-CLP film was equal to 0.6 ±0.1 mg TAE per g of film, as indicated in a previous study [32] (or 6.4± 0.7 mg TAE per g lignin in film) (Table S1). The corresponding valuesfor CNC-CLP and CNC-FA01were 3.6±0.2mg and 6.4±0.4mg TAE perg of film respectively. Interestingly, the better performance of PB1000compared to CLPwas retained after incorporation in the film, indicatingthat the nanostructure of CLP was preserved within the cellulosic film.However, film processing led to a lower RSC for both samples, with a


reduction from 3.4 (for ABTS•+) to 4.4 (for DPPH•) times for PB1000 andfrom5.0 (for ABTS•+) to 6.8 (for DPPH•) times for CLP. These results sug-gested a lower accessibility of the radicals to the phenol groups as a con-sequence of their embedding in the cellulose matrix and limitedextractability (Fig. 6 and table insert: 22% antioxidant activity in filmagainst 78% antioxidant activity released inwater after 6 h of immersionof CNC-PB1000 film and 37% and 63% for CNC-CLP, respectively). Thisphenomenon was more pronounced when the lignin samples were in-troduced into the CNF film by blending prior to high-pressure filtrationanddrying at room temperature (CNF-CLP, 10wt%). Indeed, in that case,no radical scavengingwas detected towards DPPH• in ethanolic solutionand the EC50film for ABTS•+ in aqueous solutionwas 4 times higher thanthe activity of CNC-CLP, with an identical rate of release after immersioninwater (40% antioxidant activity in film against 60% released in water)(Fig. 6 insert). This different behaviour of CNF-CLP towards DPPH• andABTS•+ was probably due to the lower swelling of the film in ethanolcompared to that in water. Greater swelling of the films could enablebetter penetration of the reagents into the film. Thus, all films preparedby the oxidative procedure showed higher RSC than films prepared byfiltration and hot pressing process, which suggested that phenolic com-pounds were less accessible with this last process. The difference by afactor approximately 4 in the ABTS•+ test between the CNF-CLP andCNC-CLP films could also be explained by the difference of film area incontact with water (84.8 mm2 (corresponding to approximately2.6 mg of film) for CNF-CLP 10 wt% and 17.5 mm2 corresponding to ap-proximately 0.7 mg of film) for CNC-CLP 10 wt%, (Table S1). Taken to-gether, these results showed that both the film preparation and thetest influenced the RSC of the films, with maximal performance ob-served for ABTS•+ in aqueous solution. Moreover, they showed thatthe CLP structure was preserved in the film, which decreased the acces-sibility of some of the phenol groups to the surrounding medium. De-spite these results, the free remaining portion of lignin in thenanocomposite film was able to trap free radicals in the medium(more in water than in ethanol) that were in contact with the materialand remains an active additive for the oxygen barrier properties, whichare observed in well-known cellulosic materials for packaging [64].

To further assess the influence of the lignin chemical structure on theRSC of the films, F1CH40 and F2CH40were incorporated in the CNC filmat 10 wt% by a mild oxidative process for comparison with PB1000 andCLP (Fig. 6). Whereas PB1000 supramolecular organization into CLP ledto reduced RSC values, particularly after film formation, its conversioninto phenol-rich oligomers (F1CH40 and F2CH40) led to enhancedRSC values, notably for DPPH•, thus leading to the following RSC-decreasing order of F1CH40 ≥ F2CH40 ≫ PB1000 ≫ CLP. As for PB1000

Fig. 6. EC50film values (g lignin. mol−1 R•) of all cellulose-based films (CNC and CNF) andantioxidant activity released after 6 h of immersion in the respective stable ethanolicDPPH• and aqueous ABTS•+ solutions (insert). F1CH40, F2CH40, PB1000 and CNC-CLPfilms were prepared by chemical oxidation procedures. CNF-CLP was prepared bypressure-assisted filtration and drying at ambient temperature.

145

and CLP, incorporation of the phenol-rich fractions in the cellulosicfilms decreased their RSC measured in solution (3-fold increase inEC50sol). However, their EC50film value was still 4.5-fold lower thanthat of unmodified PB1000-basedfilms. In addition to their initial higherphenol group content, the lower molar mass could explain the goodperformance in the films. Indeed, the lower molar mass favoured botha higher miscibility with thematrix and partial release into themedium[22]. Moreover, topographic images of the films showed fewer nodulesfor films containing F1CH40 or F2CH40 than for those containingPB1000 and CLP (Fig. 2). This lower level of supramolecular organiza-tion would account for the higher accessibility of phenol groups.

These results demonstrated that both the phenol group content andtheir accessibility to radicals control the radical scavengingperformanceof materials. The accessibility is higher when the phenol groups are lo-cated at the surface of the films and when the phenolic compoundsare not assembled in particles or nodules in the matrix. Indeed, despitethe presence ofwater-trapping cavities allowing penetration of aqueoussolution into the film, the transfer of protons between free phenolgroups and radical species requires the exposure of the phenols towardsthe surrounding medium. This conclusion was clearly supported by theabsence of RSC towards DPPH• in ethanolic solution for the CNF-CLPfilm. Even if soluble radicals are not reduced efficiently in the presenceof nanoparticles, some radicals might be trapped by the particles andstabilized through aromatic ring interactions. Moreover, the RSC of thefilms might depend on the stabilization of the phenoxy radicals formedthrough DPPH• or ABTS•+ reduction and/or after mild oxidative treat-ment with Fenton's reagent. Therefore, an investigation of the radicalstability in the films using phenoxy radicals was undertaken.

3.7. Benefit of lignin structuration to maintain the antibacterial propertiesof cellulose-lignin films

To evaluate the antimicrobial properties of PB1000 in CNC-basedfilms, an initial screening was carried out by performing inhibitiontests in agar medium. These plate tests enabled us to qualitatively eval-uate the properties of the filmswith regard to different bacterial species,namely, E. coli, P. aeruginosa, S. aureus, B. subtilis, and E. hirae. The filmsshowed significant activity against E. coli and S. aureus as the main rep-resentative gram-negative and gram-positive bacteria, respectively,which are responsible for food poisoning, localized suppurative infec-tions and, in extreme cases, potentially fatal infections (for the latter).Thus, no growth was visualized on the surface of the films and a slightinhibition zone was observed on the periphery of these films (data notshown). To quantify this antibacterial effect against these two bacteria,inhibition tests in liquid medium were carried out. For this purpose,CNC-PB1000 film sections of approximately 4 mg weight mass (i.e., 25mm2 of surface area) and from one week to a few months of age wereincubated in the respective culture media (108 CFU/mL) with respectto the control. A bacterial viability reduction was not clearly observedagainst E. coli (Fig. S2). In contrast, after 1 and 3 h of contact with theCNC-PB1000 film, the growth factor reduction of S. aureus (LogR) reached 1.0 and 1.5 from 2.0 to 3.4 for the one-week- and four-month-old films, respectively. Similar contrasting antibacterial proper-ties were observed for lignin-precipitated cellulose-lignin beads[65,66] and more recently for polyethylene blends [67]. Similar experi-ments were performed with PB1000 and CLP by using supported coat-ing films on quartz slides with the objective of increasing the surfacecontact area (3.1 cm2) with S. aureus culture media. Fig. 7 confirmsthe decreased growth of S. aureus for PB1000 in the CNC film, whichhad a Log R value between 0.3 and 0.4 after 1 and 3 h of incubation.The CLP seemed to be more efficient in the CNF film with higher Log Rvalues (between 0.35 and 0.5), which was perhaps due to the high dis-tribution of nanoparticles on the cellulosicfilm surface observedbyAFM(Fig. 2D) and the film mass difference between the CNC-lignin filmcoated on the quartz slide (thickness on the order of 1.5 μm for thecoated film and 80 μm for the CNF-lignin self-supported film). In

Fig. 7. Growth factor reduction (Log R) towards Staphylococcus aureus (gram+) of PB1000 and CLP lignin samples incorporated in CNC- and CNF-based films at 17 wt% and 10 wt%,respectively (grey bars). Controls (blue bars) are the corresponding cellulose films without lignin.


comparison, the CNC and CNF control films showed a decrease ingrowth that did not exceed 0.1 and 0.35 after similar time periods, re-spectively. Thus, regardless of the formulation mode and the nodularstructuration level of lignin oligomers, we confirmed that the ligninsamples were effective against the gram-positive bacteria S. aureus butless effective against the gram-negative bacteria E. coli. According toprevious studies, the antimicrobial activity of lignin is based on its inter-action with the surface layer of bacteria composed of peptidoglycan[66]. Moreover, cellulose films prepared with mild oxidative processesshow stable phenoxy radicals [39] that could influence antimicrobial ac-tivity as redox-active antimicrobial complexes based on diphenol due totheir possible ability to affect the electron-transport chain on bacteria[68].

3.8. Benefit of lignin structuration with respect to phenoxy radicalstabilization

The Fenton reaction involves reactive oxygen species (ROS) in mix-tures of cellulose and ligninmodel compounds (LMCs), which turn phe-nols into phenoxy radicals [39]. The formation and evolution ofpersistent phenoxy radicals in cellulosic films were demonstrated bythe proportional EPR signal to the EC50sol values and the apparent aver-agemolarmass of the LMC. To further elucidate the organization of phe-nols in CLP within cellulosic materials, an EPR analysis was performedon the CNC-CLP film under the conditions of an oxidative reaction pro-cess (1mM FeSO4/0.1 mMH2O2), and the results were compared to theCNC-PB1000 film (Fig. 8). EPR signals were observed at a g factor of2.0030, which is consistent with that of phenoxy radicals as CNC-LMCfilms [39]. The amplitude signals, which are expressed per gram of lig-nin in the films, were proportional to the apparent polymerization de-gree considering the mean mass value of the phenylpropan unit(C9H9O2(OCH3)1–2) of 196 g mol−1, and they were inversely propor-tional to the phenol content (PhOH) values measured by 31P NMR ac-cording to [22] (insert in Fig. 8). The comparison of the EPR databetween PB1000 and CLP showed that oxidation by Fenton's reagentwas responsible for the formation of phenoxy radicals in the films,with fewer radicals for PB1000 than for CLP (1988 106 a.u. g−1 filmagainst 3195 106 a.u. g−1 film). The empirical content value of the phe-nolic units (PhOHEPR, mmol g−1 lignin) converted into phenoxy radicalsper CLP can be estimated from the relationship between the EPR signalamplitude (106 g−1 lignin) and total phenol content and was on theorder of 0.77 mmol g−1 for the CNC-CLP film and 1.75 mmol g−1 forthe CNC-PB1000 film. These values corresponded to 24% and 65% ofthe total PhOH content measured by 31P NMR of raw PB1000 (Table 1and Fig. 8 insert). This finding is consistent with the nodular structuresobserved in both CNC-based films (Fig. 2C and D) and suggests thatsome phenolic groups are not free but are either linked to cellulose byhydrogen bonds or oxidized during the preparation of the films, whichcan explain their loss of antioxidant power (Fig. 6). Nevertheless, thisloss of antioxidant power of CLP can favour another functionality, such

146

as the capacity to trap stable organic radicals in composites, whichcould be useful for electronic applications. Moreover, the EPR resultscan also be used to estimate a rough amount of PB1000 oligomermoles constitutive of one CLP as follows. Considering that one CLPwould be composed of n moles of initial PB1000 oligomer (n times 7monomer units) with a respective total phenol content of 2.7 mmolg−1, the theoretical EPR signal should be equal to 3195 106 a.u. g−1 lig-nin multiplied by the ratio PhOHPB1000/PhOHEPR (= 2.70/0.77 for CNC-CLP) (Fig. SI3). This new empirical EPR signal would then correspondto the theoretical number of monomer units for one CLP (with a meansize value of 114.7± 5.2 nm) on the order of 75monomer units accord-ing to the relation between the EPR signal amplitude (10−6 g−1 lignin)and monomer unit number (Pd, polymerization degree) of LMC(Fig. SI3). This unit number corresponds to a mean of eleven oligomericPB1000moles per CLP (Table 1). In conclusion, some phenolic groups ofCLP can be converted into persistent phenoxy radicals. These groups areoriented towards the outside of the nanoparticle. However, most ofthem seem to be either localized in the core of the nanoparticle or linkedto cellulose by hydrogen bonds in the composite film.

4. Conclusions

In this study, the diversity of lignin structural levels was successfullyexploited to determine the relationships between lignin functionalgroup accessibility, cellulose-lignin composite morphologies and ligninfunctional properties in the different systems. The nodular organizationof lignin compounds in the cellulosicmatrixwas shown to be controlledby the amount of accessible phenolic groups in addition to the polymer-ization degree. They were shown to govern the capacity of the compos-ites to trap water aggregates and accounted and to scavenge solubleradical species in ethanolic or aqueous solutions. Converting lignininto phenol-rich oligomers allowed to enhance the RSC while self-assembly into CLP reduced it, as a consequence of the lower accessibilityof phenolic groups. However lignin structuration did not affect the anti-bacterial properties towards S. aureus. Thus, both CLP and lignin deriva-tives have thepotential for use as active additives in cellulosic packagingmaterials. The capacity of CLP to trap stable phenoxy radicals in cellu-lose composites after oxidative treatment offers new perspectives forsensors and electronic applications.

Funding sources

This work was funded by the Bio Based Industry Joint Undertakingunder the EuropeanUnion's Horizon 2020 research and innovation pro-gramme within the Zelcor project (under the grant number No720303), part of the COFILI project (grant number D201550245) forAFM measurements funded by the Grand Est Region and the EuropeanFEDER Programme and the Lignoxyl project for EPRmeasurements sup-ported by the Agence Nationale de la Recherche (ANR) through the Car-not Institutes 3BCAR (www.3bcar.fr) and Qualiment (https://

http://www.3bcar.fr

https://qualiment.fr/

Fig. 8. X-band EPR spectra of crushed films, CNC-PB1000 and CNC-CLP (17 wt% lignin content) prepared with Fenton's reagent (1 mM FeSO4, 0.1 mM H2O2, pH 3). Insert: Relationshipbetween the phenol group content (PhOH, mmol g−1) in LMC samples determined by 31P NMR and the phenoxy radical signal amplitude per g of lignin in CNC-based film. (The linearequation was [EPR signal 106 g−1 lignin] = −1248.2 [PhOH mmol g−1] + 4231, R2: 0.9976).


qualiment.fr/) (no. 3 no. 19-CARN-001-01 and no. 16-CARN 001-01).The EPR data in this manuscript were obtained using equipment sup-ported jointly by the French National Ministry of Research (PPF IRPE),the “Fondation pour la Recherche Médicale” (FRM DGE20061007745),and the CNRS (Department of Chemistry and Life Sciences). The IJPBbenefits from the support of the LabEx Saclay Plant Sciences-SPS(ANR-10-LABX-552 0040-SPS).

CRediT authorship contribution statement

V.A\\B., B.K., S.B. and M.Ö. conceived the scientific strategy of thestudy within the framework of the Zelcor project, which is coordinatedby S.B. E.G. and Y.M.F. performed the EPR measurements andinterpreted the experimental data. G.N.R., L.F., and B.G. conceived ofthe cellulose films and performed their spectroscopic characterizationsand antioxidant capacity measurements. M.P. performed and analysedthe assessment of the water accessibility of films. C.M. performed theAFM observations and surface topography analysis. B.G. and A.G-C. per-formed the antibacterial tests on cellulose-lignin films and interpretedthe data. B.C. and A.M. prepared the derived lignin samples by ionic liq-uid treatment and performed the 31P NMRmeasurements as D.C. on thelignin model compounds. The article was written by V.A\\B, with im-portant contributions from S.B. and M.Ö. All the authors read, reviewedand approved the final manuscript.

Acknowledgement

The authors thank Brigitte Chabbert (UMR FARE, Reims, France) andMichael Molinari (CBMN UMR CNRS 5248, Université de Bordeaux, IPB,Pessac, 33600, France) for their scientific support and fruitful discussionon lignin and AFM measurements, respectively; Dr. Raphael Coste forthe complementary surface topography analysis by AFM, and the plat-formsNano'mat, PLAneT of the University Reims-Champagne Ardennes,and the IJPB Institute Plant Observatory Chemistry andMetabolismPlat-form for access to AFM and NMR.

Author contributions section

V.A-B., B.K., S.B. and M.Ö. conceived the scientific strategy of thestudy within the framework of the Zelcor project, which is coordinated

147

by S.B., E.G. and Y.M.F. performed the EPR measurements andinterpreted the experimental data. G.N.R., L.F. and B.G. conceived ofthe cellulose films and performed their spectroscopic characterizationsand antioxidant capacity measurements. M.P. performed and analysedthe assessment of the water accessibility of films. C.M. performed theAFM observations and surface topography analysis. B.G. and A.G-C. per-formed the antibacterial tests on cellulose-lignin films.

Appendix A. Supplementary data

Supplementary data to this article can be found online at https://doi.org/10.1016/j.ijbiomac.2021.03.081.

References

[1] M.A. Hubbe, A. Ferrer, P. Tyagi, Y.Y. Yin, C. Salas, L. Pal, O.J. Rojas, Nanocellulose inthin films, coatings, and plies for packaging applications: a review, Bioresources12(1) (2017) 2143–2233.

[2] E. Sourty, D.H. Ryan, R.H. Marchessault, Characterization of magnetic membranesbased on bacterial and man-made cellulose, Cellulose 5(1) (1998) 5–17. doi:10.1023/A:1009207027869.

[3] G. Kwon, K. Lee, D. Kim, Y. Jeon, U.J. Kim, J. You, Cellulose nanocrystal-coatedTEMPO-oxidized cellulose nanofiber films for high performance all-cellulose nano-composites, J. Hazard. Mater. 398 (2020)https://doi.org/10.1016/j.jhazmat.2020.123100.

[4] F. Jiang, A.R. Esker, M. Roman, Acid-catalyzed and solvolytic desulfation of H2SO4-hydrolyzed cellulose nanocrystals, Langmuir 26 (23) (2010) 17919–17925,https://doi.org/10.1021/la1028405.

[5] D. Klemm, F. Kramer, S. Moritz, T. Lindstrom, M. Ankerfors, D. Gray, A. Dorris,Nanocelluloses: a new family of nature-based materials, Angewandte Chemie-International Edition 50 (24) (2011) 5438–5466, https://doi.org/10.1002/anie.201001273.

[6] E. Kontturi, P. Laaksonen, M.B. Linder, Nonappa, A.H. Groschel, O.J. Rojas, O. Ikkala,Advanced materials through assembly of nanocelluloses, Adv. Mater. 30(24)(2018). doi:https://doi.org/10.1002/adma.201703779.

[7] X.Z. Xu, F. Liu, L. Jiang, J.Y. Zhu, D. Haagenson, D.P. Wiesenborn, Cellulosenanocrystals vs. cellulose nanofibrils: a comparative study on their microstructuresand effects as polymer reinforcing agents, ACS Appl. Mater. Interfaces 5 (8) (2013)2999–3009, https://doi.org/10.1021/am302624t.

[8] V. Aguié-Béghin, G. Paës, M.Molinari, B. Chabbert, Films and Coatings from Lignocel-lulosic Polymers, Edible Films and Coatings: Fundamentals and Applications, in: M.Pilar Montero Garcia (Ed.), Food Preservation Technology Series, CRC PressTaylor&Francis Group 2017, p. 598.

[9] N. Mittal, F. Ansari, V.K. Gowda, C. Brouzet, P. Chen, P.T. Larsson, S.V. Roth, F. Lundell,L. Wagberg, N.A. Kotov, L.D. Soderberg, Multiscale control of nanocellulose assem-bly: transferring remarkable nanoscale fibril mechanics to macroscale fibers, ACSNano 12(7) (2018) 6378–6388. doi:https://doi.org/10.1021/acsnano.8b01084.

https://qualiment.fr/



https://doi.org/10.1016/j.jhazmat.2020.123100

https://doi.org/10.1016/j.jhazmat.2020.123100

https://doi.org/10.1021/la1028405

https://doi.org/10.1002/anie.201001273

https://doi.org/10.1002/anie.201001273

https://doi.org/10.1021/am302624t

http://refhub.elsevier.com/S0141-8130(21)00599-7/rf0025





[10] R.J. Moon, A. Martini, J. Nairn, J. Simonsen, J. Youngblood, Cellulose nanomaterialsreview: structure, properties and nanocomposites, Chem. Soc. Rev. 40 (7) (2011)3941–3994, https://doi.org/10.1039/C0CS00108B.

[11] Z.D. Zhu, S.Y. Fu, N. Lavoine, L.A. Lucia, Structural reconstruction strategies for thedesign of cellulose nanomaterials and aligned wood cellulose-based functional ma-terials - a review, Carbohydr. Polym. 247 (2020)https://doi.org/10.1016/j.carbpol.2020.116722.

[12] R. Rinaldi, R. Jastrzebski, M.T. Clough, J. Ralph, M. Kennema, P.C.A. Bruijnincx, B.M.Weckhuysen, Paving the way for lignin valorisation: recent advances in bioengi-neering, biorefining and catalysis, Angewandte Chemie-International Edition 55(29) (2016) 8164–8215, https://doi.org/10.1002/anie. 201510351.

[13] A. Grossman, W. Vermerris, Lignin-based polymers and nanomaterials, Curr. Opin.Biotechnol. 56 (2019) 112–120, https://doi.org/10.1016/j.copbio.2018.10.009.

[14] R. Singh, J.G. Hu, M.R. Regner, J.W. Round, J. Ralph, J.N. Saddler, L.D. Eltis, Enhanceddelignification of steam-pretreated poplar by a bacterial laccase, Sci. Rep. 7 (2017)https://doi.org/10.1038/srep42121.

[15] T. Tokimatsu, S. Miyata, S.H. Ahn, T. Umezawa, T. Hattori, M. Shimada, Degradationof non-phenolic beta-O-4 type lignin carbohydrate complex model compounds bylignin peroxidase from the white-rot fungus Phanerochaete chrysosporium,Mokuzai Gakkaishi 42(2) (1996) 173–179.

[16] D.S. Bajwa, G. Pourhashem, A.H. Ullah, S.G. Bajwa, A concise review of current ligninproduction, applications, products and their environmental impact, Ind. Crop. Prod.139 (2019)https://doi.org/10.1016/j.indcrop.2019.111526.

[17] S. Constant, H.L.J. Wienk, A.E. Frissen, P. de Peinder, R. Boelens, D.S. van Es, R.J.H.Grisel, B.M. Weckhuysen, W.J.J. Huijgen, R.J.A. Gosselink, P.C.A. Bruijnincx, New in-sights into the structure and composition of technical lignins: a comparative charac-terisation study, Green Chem. 18(9) (2016) 2651–2665. doi:https://doi.org/10.1039/c5gc03043a.

[18] D.L. Gall, J. Ralph, T.J. Donohue, D.R. Noguera, Biochemical transformation of ligninfor deriving valued commodities from lignocellulose, Curr. Opin. Biotechnol. 45(2017) 120–126, https://doi.org/10.1016/j.copbio.2017.02.015.

[19] T. Aro, P. Fatehi, Production and application of lignosulfonates and sulfonated lignin,Chemsuschem 10 (9) (2017) 1861–1877, https://doi.org/10.1002/cssc.201700082.

[20] M. Toikka, G. Brunow, Lignin-carbohydrate model compounds. Reactivity of methyl3-O-(alpha-L-arabinofuranosyl)-beta-D-xylopyranoside and methyl beta-D-xylopy-ranoside towards a beta-O-4-quinone methide, Journal of the Chemical Society-Perkin Transactions 1 (13) (1999) 1877–1883. doi:10.1039/a900434c.

[21] T. Tokimatsu, T. Umezawa, M. Shimada, Synthesis of four diastereomeric lignin car-bohydrate complexes (LCC) model compounds composed of a beta-0-4 ligninmodel linked to methyl beta-D-glucoside, Holzforschung 50(2) (1996) 156–160.doi:https://doi.org/10.1515/hfsg.1996.50.2.156.

[22] V. Aguié-Béghin, L. Foulon, P. Soto, D. Crônier, E. Corti, F. Legee, L. Cezard, B.Chabbert, M.N. Maillard, W.J.J. Huijgen, S. Baumberger, Use of food and packagingmodelmatrices to investigate the antioxidant properties of biorefinery grass lignins,J. Agric. Food Chem. 63 (45) (2015) 10022–10031, https://doi.org/10.1021/acs.jafc.5b03686.

[23] A. Majira, B. Godon, L. Foulon, J.C. van der Putten, L. Cezard, M. Thierry, F. Pion, A.Bado-Nilles, P. Pandard, T. Jayabalan, V. Aguié-Béghin, P.H. Ducrot, C. Lapierre, G.Marlair, R.J.A. Gosselink, S. Baumberger, B. Cottyn, Enhancing the antioxidant activ-ity of technical lignins by combining solvent fractionation and ionic-liquid treat-ment, Chemsuschem 12 (21) (2019) 4799–4809, https://doi.org/10.1002/cssc.201901916.

[24] M. Österberg, M.H. Sipponen, B.D. Mattos, O.J. Rojas, Spherical lignin particles: a re-view on their sustainability and applications, Green Chem. 22 (9) (2020)2712–2733, https://doi.org/10.1039/d0gc00096e.

[25] S. Beisl, A. Friedl, A. Miltner, Lignin from micro- to nanosize: applications, Int. J. Mol.Sci. 18 (11) (2017)https://doi.org/10.3390/ijms18112367.

[26] M. Culebras, M.J. Sanchis, A. Beaucamp, M. Carsi, B.K. Kandola, A.R. Horrocks, G.Panzetti, C. Birkinshaw, M.N. Collins, Understanding the thermal and dielectric re-sponse of organosolv and modified Kraft lignin as a carbon fibre precursor, GreenChem. 20(19) (2018). doi:https://doi.org/10.1039/c8gc01577e/rsc.li/greenchem.

[27] H.L. Dong, M. Li, Y.C. Jin, Y. Wu, C.X. Huang, J.L. Yang, Preparation of graphene-likeporous carbons with enhanced thermal conductivities from lignin nano-particlesby combining hydrothermal carbonization and pyrolysis, Frontiers in Energy Re-search 8 (2020)https://doi.org/10.3389/fenrg.2020.00148.

[28] Y. Qian, X.Q. Qiu, X.W. Zhong, D.L. Zhang, Y.H. Deng, D.J. Yang, S.P. Zhu, Lignin re-verse micelles for UV-absorbing and high mechanical performance thermoplastics,Ind. Eng. Chem. Res. 54 (48) (2015) 12025–12030, https://doi.org/10.1021/acs.iecr.5b03360.

[29] P. Ortiz-Serna, M. Carsi, M. Culebras, M.N. Collins, M.J. Sanchis, Exploring the role oflignin structure in molecular dynamics of lignin/bio-derived thermoplastic elasto-mer polyurethane blends, Int. J. Biol. Macromol. 158 (2020) 1369–1379, https://doi.org/10.1016/j.ijbiomac.2020.04.261.

[30] A. Hambardzumyan, L. Foulon, N.B. Bercu, M. Pernes, J.E. Maigret, M. Molinari, B.Chabbert, V. Aguié-Béghin, Organosolv lignin as natural grafting additive to improvethe water resistance of films using cellulose nanocrystals, Chem. Eng. J. 264 (2015)780–788, https://doi.org/10.1016/j.cej.2014.12.004.

[31] A. Hambardzumyan, L. Foulon, B. Chabbert, V. Aguié-Béghin, Natural organic UV-absorbent coatings based on cellulose and lignin: designed effects on spectroscopicproperties, Biomacromolecules 13 (12) (2012) 4081–4088, https://doi.org/10.1021/bm301373b.

[32] M. Farooq, T. Zou, G. Rivière, M.H. Sipponen, M. Österberg, Strong, ductile, and wa-terproof cellulose nanofibril composite films with colloidal lignin particles,Biomacromolecules 20 (2) (2019) 693–704, https://doi.org/10.1021/acs.biomac.8b01364.

148

[33] A.P. Richter, J.S. Brown, B. Bharti, A. Wang, S. Gangwal, K. Houck, E.A.C. Hubal, V.N.Paunov, S.D. Stoyanov, O.D. Velev, An environmentally benign antimicrobial nano-particle based on a silver-infused lignin core, Nat. Nanotechnol. 10(9) (2015) 817doi:https://doi.org/10.1038/NNANO.2015.141.

[34] G.N. Rivière, A. Korpi, M.H. Sipponen, T. Zou, M.A. Kostiainen, M. Österberg, Agglom-eration of viruses by cationic lignin particles for facilitated water purification, ACSSustain. Chem. Eng. 8 (10) (2020) 4167–4177, https://doi.org/10.1021/acssuschemeng.9b06915.

[35] K. Kanomata, N. Fukuda, T. Miyata, P.Y. Lam, T. Takano, Y. Tobimatsu, T. Kitaoka,Lignin-inspired surface modification of nanocellulose by enzyme-catalyzed radicalcoupling of coniferyl alcohol in pickering emulsion, ACS Sustain. Chem. Eng. 8 (2)(2020) 1185–1194, https://doi.org/10.1021/acssuschemeng.9b06291.

[36] L. Bai, L.G. Greca, W.C. Xiang, J. Lehtonen, S.Q. Huan, R.W.N. Nugroho, B.L. Tardy, O.J.Rojas, Adsorption and assembly of cellulosic and lignin colloids at oil/water inter-faces, Langmuir 35 (3) (2019) 571–588, https://doi.org/10.1021/acs.langmuir.8b01288.

[37] M.H. Sipponen, M. Smyth, T. Leskinen, L.S. Johansson, M. Österberg, All-lignin ap-proach to prepare cationic colloidal lignin particles: stabilization of durable Picker-ing emulsions, Green Chem. 19 (24) (2017) 5831–5840, https://doi.org/10.1039/c7gc02900d.

[38] C.X. Huang, J.M. Ma, W.Y. Zhang, G. Huang, Q. Yong, Preparation of lignosulfonatesfrom biorefinery lignins by sulfomethylation and their application as a water re-ducer for concrete, Polymers 10 (8) (2018)https://doi.org/10.3390/polym10080841.

[39] E. Gerbin, Y.M. Frapart, C. Marcuello, B. Cottyn, L. Foulon, M. Pernes, D. Crônier, M.Molinari, B. Chabbert, P.H. Ducrot, S. Baumberger, V. Aguié-Béghin, B. Kurek, Dualantioxidant properties and organic radical stabilization in cellulose nanocompositefilms functionalized by in situ polymerization of coniferyl alcohol,Biomacromolecules 21 (8) (2020) 3163–3175, https://doi.org/10.1021/acs.biomac.0c00583.

[40] L. Zheng, P. Yu, Y. Zhang, P. Wang, W. Yan, B. Guo, C. Huang, Q. Jiang, Evaluating thebio-application of biomacromolecule of lignin-carbohydrate complexes (LCC) fromwheat straw in bone metabolism via ROS scavenging, Int. J. Biol. Macromol. 176(2021) 13–25, https://doi.org/10.1016/j.ijbiomac.2021.01.103.

[41] J.H. Lora, Lignin: a platform for renewable aromatic polymericmaterials, Quality Liv-ing through Chemurgy and Green Chemistry2016, pp. 221–261.

[42] M. Österberg, J. Vartiainen, J. Lucenius, U. Hippi, J. Seppala, R. Serimaa, J. Laine, A fastmethod to produce strong NFC films as a platform for barrier and functional mate-rials, ACS Appl. Mater. Interfaces 5 (11) (2013) 4640–4647, https://doi.org/10.1021/am401046x.

[43] C. Marcuello, L. Foulon, B. Chabbert, M. Molinari, V. Aguié-Béghin, Langmuir-Blodgett procedure to precisely control the coverage of functionalized AFM cantile-vers for SMFS measurements: application with cellulose nanocrystals, Langmuir 34(32) (2018) 9376–9386. doi:10.1021/acs.langmuir.8b01892.

[44] A. Granata, D.S. Argyropoulos, 2-Chloro-4,4,5,5-tetramethyl-1,3,2-dioxaphospholane, a reagent for the accurate determination of the uncondensedand condensed phenolic moieties in lignins, J. Agric. Food Chem. 43 (6) (1995)1538–1544, https://doi.org/10.1021/jf00054a023.

[45] J.E. Sader, I. Larson, P. Mulvaney, L.R. White, Method for the calibration of atomic-force microscope cantilevers, Rev. Sci. Instrum. 66 (7) (1995) 3789–3798, https://doi.org/10.1063/1.1145439.

[46] K. Meinander, T.N. Jensen, S.B. Simonsen, S. Helveg, J.V. Lauritsen, Quantification oftip-broadening in non-contact atomic force microscopy with carbon nanotubetips, Nanotechnology 23 (40) (2012)https://doi.org/10.1088/0957-4484/23/40/405705.

[47] L. Muraille, M. Pernes, A. Habrant, R. Serimaa, M. Molinari, V. Aguié-Béghin, B.Chabbert, Impact of lignin on water sorption properties of bioinspired self-assemblies of lignocellulosic polymers, Eur. Polym. J. 64 (2015) 21–35, https://doi.org/10.1016/j.eurpolymj.2014.11.040.

[48] W. Brand-Williams, M.E. Cuvelier, C. Berset, Use of a free radical method to evaluateantioxidant activity, Lebensm.-Wiss. u.-Technol. 28 (1995) 25–30.

[49] R. Re, N. Pellegrini, A. Proteggente, A. Pannala, M. Yang, C. Rice-Evans, Antioxidantactivity applying an improved ABTS radical cation decolorization assay, Free Radic.Biol. Med. 26 (9–10) (1999) 1231–1237, https://doi.org/10.1016/S0891-5849(98)00315-3.

[50] N. Pellegrini, F. Visioli, S. Buratti, F. Brighenti, Direct analysis of total antioxidant ac-tivity of olive oil and studies on the influence of heating, J. Agric. Food Chem. 49 (5)(2001) 2532–2538, https://doi.org/10.1021/jf001418j.

[51] V. Aguié-Béghin, S. Baumberger, B. Monties, R. Douillard, Formation and characteri-zation of spread ligin layers at the air water interface, Langmuir 18 (2002)5190–5196, https://doi.org/10.1021/la011766v.

[52] I. Boukari, J.L. Putaux, B. Cathala, A. Barakat, B. Saake, C. Remond, M. O’Donohue, B.Chabbert, In vitro model assemblies to study the impact of lignin-carbohydrate in-teractions on the enzymatic conversion of xylan, Biomacromolecules 10 (9)(2009) 2489–2498, https://doi.org/10.1021/bm9004518.

[53] T. Leskinen, M. Smyth, Y. Xiao, K. Lintinen, M.L. Mattinen, M.A. Kostiainen, P. Oinas,M. Österberg, Scaling up production of colloidal lignin particles, Nordic Pulp & PaperResearch Journal 32(4) (2017) 586–596. doi:10.3183/NPPRJ-2017-32-04-p586-596.

[54] C. Marcuello, L. Foulon, B. Chabbert, V. Aguié-Béghin, M. Molinari, Atomic force mi-croscopy reveals how relative humidity impacts the Young’s modulus of lignocellu-losic polymers and their adhesion with cellulose nanocrystals at the nanoscale, Int. J.Biol. Macromol. 147 (2020) 1064–1075, https://doi.org/10.1016/j.ijbiomac.2019.10.074.

[55] Z.D. Zhu, S.Y. Fu, L.A. Lucia, Tuning themorphology ofmicroparticles from spray dry-ing of cellulose nanocrystal suspensions by hydrophobic lignin, ACS Sustain. Chem.Eng. 7 (5) (2019) 5376–5384, https://doi.org/10.1021/acssuschemeng.8b06399.

https://doi.org/10.1039/C0CS00108B

https://doi.org/10.1016/j.carbpol.2020.116722

https://doi.org/10.1016/j.carbpol.2020.116722

https://doi.org/10.1002/anie. 201510351

https://doi.org/10.1016/j.copbio.2018.10.009

https://doi.org/10.1038/srep42121

https://doi.org/10.1016/j.indcrop.2019.111526

https://doi.org/10.1016/j.copbio.2017.02.015

https://doi.org/10.1002/cssc.201700082

https://doi.org/10.1021/acs.jafc.5b03686

https://doi.org/10.1021/acs.jafc.5b03686



https://doi.org/10.1039/d0gc00096e

https://doi.org/10.3390/ijms18112367

https://doi.org/10.3389/fenrg.2020.00148

https://doi.org/10.1021/acs.iecr.5b03360

https://doi.org/10.1021/acs.iecr.5b03360



https://doi.org/10.1016/j.cej.2014.12.004

https://doi.org/10.1021/bm301373b


https://doi.org/10.1021/acs.biomac.8b01364

https://doi.org/10.1021/acs.biomac.8b01364

https://doi.org/10.1021/acssuschemeng.9b06915



https://doi.org/10.1021/acs.langmuir.8b01288

https://doi.org/10.1021/acs.langmuir.8b01288

https://doi.org/10.1039/c7gc02900d

https://doi.org/10.1039/c7gc02900d

https://doi.org/10.3390/polym10080841

https://doi.org/10.3390/polym10080841

https://doi.org/10.1021/acs.biomac.0c00583

https://doi.org/10.1021/acs.biomac.0c00583


https://doi.org/10.1021/am401046x

https://doi.org/10.1021/am401046x

https://doi.org/10.1021/jf00054a023

https://doi.org/10.1063/1.1145439

https://doi.org/10.1063/1.1145439

https://doi.org/10.1088/0957-4484/23/40/405705

https://doi.org/10.1088/0957-4484/23/40/405705

https://doi.org/10.1016/j.eurpolymj.2014.11.040

https://doi.org/10.1016/j.eurpolymj.2014.11.040

https://doi.org/10.1016/S0891-5849(98)00315-3

https://doi.org/10.1016/S0891-5849(98)00315-3

https://doi.org/10.1021/jf001418j

https://doi.org/10.1021/la011766v

https://doi.org/10.1021/bm9004518





[56] M. Montero-Munoz, J.E. Ramos-Ibarra, J.E. Rodriguez-Paez, G.E. Marques, M.D.Teodoro, J.A.H. Coaquira, Growth and formation mechanism of shape-selectivepreparation of ZnO structures: correlation of structural, vibrational and opticalproperties, Phys. Chem. Chem. Phys. 22 (14) (2020) 7329–7339, https://doi.org/10.1039/c9cp06744b.

[57] M.R. Benzigar, S. Joseph, G. Saianand, A.I. Gopalan, S. Sarkar, S. Srinivasan, D.H. Park,S. Kim, S.N. Talapaneni, K. Ramadass, A. Vinu, Highly ordered iron oxide-mesoporous fullerene nanocomposites for oxygen reduction reaction andsupercapacitor applications, Microporous Mesoporous Mater. 285 (2019) 21–31,https://doi.org/10.1016/j.micromeso.2019.04.071.

[58] C.X. Huang, H.L. Dong, Z.P. Zhang, H.Y. Bian, Q. Yong, Procuring the nano-scale ligninin prehydrolyzate as ingredient to prepare cellulose nanofibril composite film withmultiple functions, Cellulose 27 (16) (2020) 9355–9370, https://doi.org/10.1007/s10570-020-03427-9.

[59] N. Gierlinger, L. Goswami, M. Schmidt, I. Burgert, C. Coutand, T. Rogge, M.Schwanninger, In situ FT-IR microscopic study on enzymatic treatment of poplarwood cross-sections, Biomacromolecules 9 (8) (2008) 2194–2201, https://doi.org/10.1021/bm800300b.

[60] M. Schwanninger, J.C. Rodrigues, H. Pereira, B. Hinterstoisser, Effects of short-timevibratory ball milling on the shape of FT-IR spectra of wood and cellulose, Vib.Spectrosc. 36 (1) (2004) 23–40, https://doi.org/10.1016/j.vibspec.2004.02.003.

[61] I.V. Pylypchuk, P.A. Linden, M.E. Lindstrom, O. Sevastyanova, New insight into thesurface structure of lignin nanoparticles revealed by H-1 liquid-state NMR spectros-copy, ACS Sustain. Chem. Eng. 8 (36) (2020) 13805–13812, https://doi.org/10.1021/acssuschemeng.0c05119.

149

[62] O. Faix, D.S. Argyropoulos, D. Robert, V. Neirinck, Determination of hydroxyl-groupsin lignins evaluation of H-1-Nmr, C-13-Nmr, P-31-Nmr, Ftir and wet chemicalmethods, Holzforschung 48(5) (1994) 387–394. doi:https://doi.org/10.1515/hfsg.1994.48.5.387.

[63] K.M. Schaich, X. Tian, J. Xie, Hurdles and pitfalls in measuring antioxidant efficacy: acritical evaluation of ABTS, DPPH, and ORAC assays (reprinted), J. Funct. Foods 18(2015) 782–796. doi:10.1016/j.jff.2015.05.024.

[64] S.S. Nair, J.Y. Zhu, Y.L. Deng, A.J. Ragauskas, Hydrogels prepared from cross-linkednanofibrillated cellulose, ACS Sustain. Chem. Eng. 2 (4) (2014) 772–780, https://doi.org/10.1021/sc400445t.

[65] K. Gabov, T. Oja, T. Deguchi, A. Fallarero, P. Fardim, Preparation, characterization andantimicrobial application of hybrid cellulose-lignin beads, Cellulose 24 (2) (2017)641–658.

[66] J. Wang, W. Vermerris, Antimicrobial nanomaterials derived from natural products-a review, Materials 9(4) (2016). doi:10.1007/s10570-016-1172-y.

[67] J. Vachon, D. Assad-Alkhateb, S. Baumberger, J. van Haveren, R.J.A. Gosselink,Monedero, M., J.M. Bermudez, Use of lignin as additive in polyethylene for food pro-tection: insect repelling effect of an ethyl acetate phenolic extract, Composites PartC.: Open Access 2 (2020) 100044 (2020) 1–9. doi:10.1016/j.jcomc.2020.100044.

[68] N.V. Loginova, H.I. Harbatsevich, N.P. Osipovich, G.A. Ksendzova, T.V. Koval’chuk, G.I.Polozov, Metal complexes as promising agents for biomedical applications,Curr. Med. Chem. 27 (31) (2020) 5213–5249, https://doi.org/10.2174/0929867326666190417143533.

https://doi.org/10.1039/c9cp06744b

https://doi.org/10.1039/c9cp06744b

https://doi.org/10.1016/j.micromeso.2019.04.071

https://doi.org/10.1007/s10570-020-03427-9

https://doi.org/10.1007/s10570-020-03427-9



https://doi.org/10.1016/j.vibspec.2004.02.003

https://doi.org/10.1021/acssuschemeng.0c05119

https://doi.org/10.1021/acssuschemeng.0c05119

https://doi.org/10.1021/sc400445t

https://doi.org/10.1021/sc400445t




https://doi.org/10.2174/0929867326666190417143533

https://doi.org/10.2174/0929867326666190417143533

tuning the functional properties of lignocellulosic films

Documents